Performance of Radial Basis Function and Support Vector Machine in time series forecasting

Author

Mamat, Mazlina ; Samad, Salina Abdul

Author_Institution

Inst. of Microengineering & Nanotechnol., Univ. Kebangsaan Malaysia, Bangi, Malaysia

fYear

2010

fDate

15-17 June 2010

Firstpage

1

Lastpage

4

Abstract

This paper compares the performance of Radial Basis Function and Support Vector Regression in time series forecasting. Both methods were trained to produce one step ahead forecasting on two chaotic time series data: Mackey Glass and Set A data from Santa Fe Competition. The criterions for comparison are based on the coefficient of determination (R2) and Root Mean Square Error (RMSE) between actual and forecasted output. Results show that SVR outperformed RBF significantly on both data particularly on Set A data.

Keywords

forecasting theory; mean square error methods; radial basis function networks; regression analysis; support vector machines; time series; Mackey Glass data forecasting; Set A data forecasting; chaotic time series data forecasting; determination coefficient; radial basis function; root mean square error; support vector machine; support vector regression; Artificial neural networks; Forecasting; Glass; Mathematical model; Support vector machines; Time series analysis; Training;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent and Advanced Systems (ICIAS), 2010 International Conference on

Conference_Location

Kuala Lumpur, Malaysia

Print_ISBN

978-1-4244-6623-8

Type

conf

DOI

10.1109/ICIAS.2010.5716201

Filename

5716201